Impaired phonological processing and abnormal perception of human voice are two critical, yet understudied, aspects of language and social impairments in children with autism spectrum disorders (ASD). Despite the prevalence and adverse impact of these deficits, the brain mechanisms underlying these phenomena have received surprisingly little experimental investigation, particularly in children with ASD. The primary goal of the proposed research is to further our understanding of basic auditory function underlying decoding of phonological content (what is being said) and speaker identity (who is saying it) in children with ASD, compared to typically developing (TD) children matched on age and language ability. To achieve this goal, we propose three experiments in which novel multivariate pattern recognition techniques will be used to increase the sensitivity for detecting fine-grained neural representations of information processing in the auditory system of children with ASD. In the first experiment, functional MRI will be used to examine discrimination of minimal pair nonsense words in auditory cortex of children with high-functioning autism (HFA) and TD children. Next, we will assess the integrity of voice-selective cortex in the superior temporal sulcus of children with HFA and TD children by quantifying brain-based discrimination of speech and non-speech environmental sounds measured with functional MRI. In the third experiment, we will use functional MRI to distinguish brain responses to words produced by each child's mother and investigate specialized brain circuits for typical and atypical processing of this salient biological signal. Findings from these experiments will provide novel information about the integrity of acoustical and phonological processing of mother's and unfamiliar voice in children with ASD, knowledge that is essential for a more complete characterization of the pervasive language deficits in autism. The proposed studies are also innovative in that they will use novel multivariate pattern recognition techniques to increase the sensitivity for detecting fine-grained neural representations of auditory information processing in children with ASD. Our research has the potential to not only provide new information about impaired speech and voice processing in autism but also, more generally, to push methodological boundaries for studying neurodevelopmental disorders. PUBLIC HEALTH RELEVANCE: The long-term goal of our research is to further our understanding of basic auditory function in individuals with autism. The primary goal of the proposed work is to better understand the neural basis of auditory information processing deficits in children with autism. Children with autism often exhibit receptive language impairments, including atypical processing of phonemes and voices, including their own mother's voice. Our goal is to better understand the neural bases of these phenomena using multivariate pattern recognition techniques applied to functional magnetic resonance imaging. A better understanding of neurobiological and behavioral bases of phonological disorders and voice recognition will provide novel insights into language deficits in autism.